City Research Online

A Multilevel Measurement Model of Social Cohesion

Bottoni, G. (2016). A Multilevel Measurement Model of Social Cohesion. Social Indicators Research, 136(3), pp. 835-857. doi: 10.1007/s11205-016-1470-7

Text - Draft Version
Download (1MB) | Preview


In spite of its currency both in academic research and political rhetoric, there are numerous attempts to define and conceptualize the social cohesion concept but there has been paid little attention to provide a rigorous and empirically tested definition. There are even fewer studies that address social cohesion in a framework of cross-cultural validation of the indicators testing the equivalence of the factorial structure across countries. Finally, as far as we know there is no study that attempt to provide an empirically tested multilevel definition of social cohesion specifying a Multilevel Structural Equation Model. This study aims to cover this gap. First, we provide a theoretical construct of social cohesion taking into account not only its multidimensionality but also its multilevel structure. In the second step, to test the validity of this theoretical construct, we perform a multilevel confirmatory factor analysis in order to verify if the conceptual structure suggested in first step holds. In addition, we test the cross-level structural equivalence and the measurement invariance of the model in order to verify if the same multilevel model of social cohesion holds across the 29 countries analysed. In the final step, we specify a second-order multilevel CFA model in order to identify the existence of a general factor that can be called “social cohesion” operating in society that accounts for the surface phenomena that we observe.

Item Type: Article
Additional Information: This is a post-peer-review, pre-copyedit version of an article published in 'Social Indicators Research'. The final authenticated version is available online at:,
Uncontrolled Keywords: Multilevel analysis Multilevel SEM Social cohesion Construct validation Multilevel measurement invariance
Divisions: School of Social Sciences > Department of Sociology

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics